How to Select a Good Training-data Subset for Transcription: Submodular Active Selection for Sequences
Abstract
Given a large un-transcribed corpus of speech utterances, we address the problem of how to select a good subset for word-level transcription under a given fixed transcription budget. We employ submodular active selection on a Fisher-kernel based graph over un-transcribed utterances. The selection is theoretically guaranteed to be near-optimal. Moreover, our approach is able to bootstrap without requiring any initial transcribed data, whereas traditional approaches rely heavily on the quality of an initial model trained on some labeled data. Our experiments on phone recognition show that our approach outperforms both average-case random selection and uncertainty sampling significantly.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2009
- Accession Number
- ADA518795
Entities
People
- Hui Lin
- Jeff Bilmes
Organizations
- University of Washington